Optical systems generally require the precise and rigid alignment of lenses, prisms, mirrors, and other optical components. Proper alignment is especially important in laser systems, where misalignment of the optical components can degrade performance. In addition, frequency doubling and other nonlinear processes involving crystals often require that the crystal be precisely aligned in order to achieve the optimum conversion efficiency to meet stringent beam performance requirements.
Stable alignment is therefore especially critical for both intracavity and extracavity, nonlinear crystal resonator configurations, in which the crystal may or may not reside within the resonator.
Alignment problems are considerably aggravated when the laser system is subjected to vibrations. Temperature cycling presents problems as well, since the optical components may expand and contract with changes in temperature at different rates. To minimize alignment problems, specialized optical mounts are frequently used to secure optical components.
In a wavelength-converted laser system, laser radiation undergoes a nonlinear optical process in some nonlinear medium, such as a nonlinear optical crystal. The nonlinear optical process converts some portion of the laser radiation to a different wavelength. The phase matching of a nonlinear crystal is typically adjusted by (1) precise cuts on the crystalline axis, (2) precise mounting of the crystal, (3) controlling the temperature of the crystal. The crystal is typically mounted to a specially-designed oven and the temperature of the crystal is adjusted by adjusting the temperature of the oven. An example of a wavelength-converted laser system is disclosed in U.S. Pat. No. 8,422,119, which is incorporated herein by reference.
Examples of non-linear crystals include, but are not limited to, lithium niobate (LiNbO3), lithium triborate (LBO), beta-barium borate (BBO), cesium lithium borate (CLBO), lithium tantalite, stoichiometric lithium tantalite (SLT) potassium titanyl phosphate (KTiOPO4 also known as KTP), ammonium dihydrogen arsenate (ADA), ammonium dihydrogen phosphate (ADP), cesium triborate (CsB3O5 or CBO), deuterated ammonium dihydrogen arsenate (DADA), deuterated ammonium dihydrogen phosphate (DADP), deuterated arginine phosphate (DLAP), rubidium di-deuterium phosphate (RbD2PO4 or DRDP, potassium aluminum borate (KABO), potassium dihydrogen arsenate (KDA), potassium dihydrogen phosphate (KDP), deuterated potassium dihydrogen phosphate (KD2PO4 or DKDP), Li2B4O7 (LB4), or lithium formate monohydrate (LFM) and isomorphs thereof, periodically poled materials such as periodically poled lithium niobate (PPLN), periodically poled lithium tantalite and periodically poled stoichiometric lithium tantalite (PPSLT).
Lithium Triborate LiB3O5 or LBO is an example of an interesting and useful nonlinear optical crystal. LBO is unique in many aspects, especially its wide transparency range, moderately high nonlinear coupling, high damage threshold and good chemical and mechanical properties. LBO crystal is also phase-matchable for second harmonic generation (SHG) and third harmonic generation (THG) of Nd:YAG and Nd:YLF lasers by using either type I or type II interaction. For SHG at room temperature, type I phase-matching can be reached and has maximum effective SHG coefficient in the principal XY and XZ planes in a wide wavelength range from 551 nm to about 3000 nm. LBO's transmission range is from 0.21 μm to 2.3 μm. LBO allows temperature-controllable non-critical phase-matching (NCPM) for 1.0-1.3 μm, Type I SHG, and also provides room temperature non-critical phase matching (NCPM) for Type II SHG at 0.8-1.1 μm. LBO is also a desirable nonlinear optical material because it possesses a reasonable angular acceptance bandwidth, reducing the beam quality requirements for source lasers.
SHG conversion efficiencies of more than 70% have been observed with LBO for pulsed Nd:YAG lasers and 30% conversion efficiencies have been observed with LBO for continuous wave (cw) Nd:YAG lasers. THG conversion efficiency of over 60% for pulse Nd:YAG lasers have been observed with LBO. LBO is also an excellent nonlinear optical (NLO) crystal for an optical parametric oscillator (OPO) or optical parametric amplifier (OPA) with a widely tunable wavelength range and high output power. Thus, LBO is a desirable non-linear optical crystal for many applications.
However, LBO is a difficult material to work with. LBO is hygroscopic and expensive. In an optical system, the LBO crystal needs to be clean, stable, e.g., perfectly still. Typically one must control the temperature of the crystal to within 0.1° C. for critical phase matching. Noncritical phase matching has much looser temperature tolerance. In addition, the mounting of the crystal is critical due to the unusually anisotropic thermal expansion of LBO. In particular, LBO has coefficients of thermal expansion of 10.8×10−5/K, −8.8×10−5/K, and 3.4×10−5/K for its x, y and z crystal axes, respectively. Optical considerations determine the cut of the crystal, i.e., phase matching. For example, a second harmonic generation (SHG) cut for LBO is easier to implement than a third harmonic generation (THG) cut. Likewise, a mounting system for SHG is easier than for THG.
The properties of LBO make it particularly difficult to mount in an oven. In the past, laser systems have used glue or a clamping mechanism (e.g., spring loads) to secure an LBO crystal to the oven for SHG or THG. Other systems have used gold flash and solder to mount LBO crystals. To avoid damage to the crystal due to anisotropic thermal expansion, a small dot of glue may be used to mount the LBO (5-mm to 15-mm long) crystal to the oven. To reduce strain, the glue is then typically cured near room temperature. However, a single dot of glue may not be sufficient to hold the LBO crystal securely and stably while protecting the crystal from chipping or cracking. Another problem is that the thermally anisotropic LBO is usually glued to a thermally isotropic metal. A mismatch in coefficients of thermal expansion (CTE) between the LBO and the metal results in differences in thermal expansion that often break the LBO crystal. Furthermore, methods involving adhesives such as glue or solder and/or mechanical clamping have significant drawbacks such as crystal chipping and cracking, or mechanical instability.
There are many existing designs for laser based crystal holding. An example of a crystal holding apparatus for laser based systems is disclosed in U.S. Pat. No. 8,305,680, the entire contents of which are incorporated herein by reference. However, no particular design successfully meets all of the ideal requirements of such an assembly, including but not limited to: cleanliness, complexity, assembly/cycle time, and re-workability.
One problem regarding current designs is cleanliness, particularly with the presence of an affixing adhesive. For a variety of reasons, some designs use glue to fix the crystal in place rather than a mechanical holding approach (See FIG. 3A). The problem is that such adhesives have the potential to outgas or degrade when exposed to stray or scattered light. This outgassing/degradation contaminates the laser optics, potentially reducing the system lifetime. It is generally recognized that removing adhesives reduces the possibility of outgassing. Therefore, although adhesive works well for fixing the crystal, it is not ideal for the overall laser longevity.
Complexity is another category that defines a successful crystal enclosure design. Complex designs are disadvantaged by cost, BOM control, and packaging size. Despite engineering a small crystal enclosure, the various parts hold significant material and management costs. Generally an assembly containing fewer parts is lower in cost. FIG. 2C shows the inherent complexity of what is considered an industry accepted crystal enclosure (16 parts). Finally, due to the small size of these enclosures, designs with more parts tend to have lots of small parts that are very difficult for a trained technician to handle.
One of the most important areas associated with crystal enclosures is the assembly/cycle time. The assembly time is the actual amount of time that it takes a technician to build the assembly. The cycle time is the assembly time plus whatever additional time passes until the assembly is ready to be put into production. For example, one downside to the glued design (shown in FIG. 3A) is the cycle time. While the actual assembly time is less than 30 minutes, there is at least 18 hours of cycle time consisting of waiting for the glue to cure before the assembly can be used. Conversely, a complex assembly (e.g., as in FIGS. 2A-2C) requires a tremendous amount of assembly time (30-60 minutes). Although there is additional cycle time involved (like the glue design), the amount of assembly time required is substantial. Finally, extended assembly and cycle times come with significant risk of damage to the sensitive optical crystals.
Another issue facing a successful design is the ability for re-work. One major disadvantage of the glue design is that it cannot be re-worked once assembled. This means that the crystal as well as the aluminum base piece cannot be recovered if repair is required. For a complex design such as the industry accepted standard (e.g., FIGS. 2A-2C), there is danger in re-working such an assembly within the laser head. It is very easy to drop these small parts, requiring the entire assembly to be removed from the laser just for crystal replacement. This is a tedious process, which adds a tremendous amount of time and effort which could be avoided if an ideal enclosure design was used instead.
It is within this context that embodiments of the present invention arise.